- Physical Review Journals
All JournalsPhysics Magazine

Physical Review B

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Letter
  • Open Access

Heat flux for semilocal machine-learning potentials

Marcel F. Langer1,2,3,*, Florian Knoop4,3,†, Christian Carbogno3, Matthias Scheffler3, and Matthias Rupp3,5,6

  • 1Machine Learning Group, Technische Universität Berlin, 10587 Berlin, Germany
  • 2Berlin Institute for the Foundations of Learning and Data, 10623 Berlin, Germany
  • 3The NOMAD Laboratory at the FHI of the Max-Planck-Gesellschaft and IRIS Adlershof of the Humboldt Universität zu Berlin, 14195 Berlin, Germany
  • 4Theoretical Physics Division, Department of Physics, Chemistry and Biology (IFM), Linköping University, 581 83 Linköping, Sweden
  • 5Department of Computer and Information Science, University of Konstanz, 78464 Konstanz, Germany
  • 6Materials Research and Technology Department, Luxembourg Institute of Science and Technology, Belvaux, Luxembourg

  • *Corresponding author: mail@marcel.science
  • †Corresponding author: florian.knoop@liu.se
PDF

Phys. Rev. B 108, L100302 – Published 13 September, 2023

DOI: https://doi.org/10.1103/PhysRevB.108.L100302

Abstract

The Green-Kubo (GK) method is a rigorous framework for heat transport simulations in materials. However, it requires an accurate description of the potential-energy surface and carefully converged statistics. Machine-learning potentials can achieve the accuracy of first-principles simulations while allowing to reach well beyond their simulation time and length scales at a fraction of the cost. In this Letter, we explain how to apply the GK approach to the recent class of message-passing machine-learning potentials, which iteratively consider semilocal interactions beyond the initial interaction cutoff. We derive an adapted heat flux formulation that can be implemented using automatic differentiation without compromising computational efficiency. The approach is demonstrated and validated by calculating the thermal conductivity of zirconium dioxide across temperatures.

View figure in article

Physics Subject Headings (PhySH)

Article Text

Supplemental Material

References (98)

  1. A. Evans, D. Clarke, and C. Levi, The influence of oxides on the performance of advanced gas turbines, J. Eur. Ceram. Soc. 28, 1405 (2008).
  2. G. J. Snyder and E. S. Toberer, Complex thermoelectric materials, Nat. Mater. 7, 105 (2008).
  3. L.-D. Zhao, S.-H. Lo, Y. Zhang, H. Sun, G. Tan, C. Uher, C. Wolverton, V. P. Dravid, and M. G. Kanatzidis, Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals, Nature (London) 508, 373 (2014).
  4. F. Knoop, T. A. R. Purcell, M. Scheffler, and C. Carbogno, Anharmonicity measure for materials, Phys. Rev. Mater. 4, 083809 (2020).
  5. L. Onsager, Reciprocal relations in irreversible processes. I, Phys. Rev. 37, 405 (1931).
  6. L. Onsager, Reciprocal relations in irreversible processes. II, Phys. Rev. 38, 2265 (1931).
  7. M. S. Green, Markoff random processes and the statistical mechanics of time-dependent phenomena, J. Chem. Phys. 20, 1281 (1952).
  8. R. Kubo, Statistical-mechanical theory of irreversible processes. I. General theory and simple applications to magnetic and conduction problems, J. Phys. Soc. Jpn. 12, 570 (1957).
  9. R. Kubo, M. Yokota, and S. Nakajima, Statistical-mechanical theory of irreversible processes. II. Response to thermal disturbance, J. Phys. Soc. Jpn. 12, 1203 (1957).
  10. A. M. Teale et al., DFT exchange: Sharing perspectives on the workhorse of quantum chemistry and materials science, Phys. Chem. Chem. Phys. 24, 28700 (2022).
  11. A. Marcolongo, P. Umari, and S. Baroni, Microscopic theory and quantum simulation of atomic heat transport, Nat. Phys. 12, 80 (2016).
  12. C. Carbogno, R. Ramprasad, and M. Scheffler, Ab Initio Green-Kubo Approach for the Thermal Conductivity of Solids, Phys. Rev. Lett. 118, 175901 (2017).
  13. A. Marcolongo, L. Ercole, and S. Baroni, Gauge fixing for heat-transport simulations, J. Chem. Theory Comput. 16, 3352 (2020).
  14. F. Knoop, M. Scheffler, and C. Carbogno, Ab initio Green-Kubo simulations of heat transport in solids: Method and implementation, Phys. Rev. B 107, 224304 (2023).
  15. M. González, Force fields and molecular dynamics simulations, J. Neutronique 12, 169 (2011).
  16. J. B. Witkoskie and D. J. Doren, Neural network models of potential energy surfaces: Prototypical examples, J. Chem. Theory Comput. 1, 14 (2005).
  17. S. Lorenz, A. Groß, and M. Scheffler, Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks, Chem. Phys. Lett. 395, 210 (2004).
  18. J. Behler and M. Parrinello, Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces, Phys. Rev. Lett. 98, 146401 (2007).
  19. A. P. Bartók, M. C. Payne, R. Kondor, and G. Csányi, Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons, Phys. Rev. Lett. 104, 136403 (2010).
  20. O. T. Unke, S. Chmiela, H. E. Sauceda, M. Gastegger, I. Poltavsky, K. T. Schütt, A. Tkatchenko, and K.-R. Müller, Machine learning force fields, Chem. Rev. 121, 10142 (2021).
  21. I. Poltavsky and A. Tkatchenko, Machine learning force fields: Recent advances and remaining challenges, J. Phys. Chem. Lett. 12, 6551 (2021).
  22. E. Prodan and W. Kohn, Nearsightedness of electronic matter, Proc. Natl. Acad. Sci. USA 102, 11635 (2005).
  23. A. K. Rappe and W. A. Goddard, Charge equilibration for molecular dynamics simulations, J. Phys. Chem. 95, 3358 (1991).
  24. T. A. Halgren and W. Damm, Polarizable force fields, Curr. Opin. Struct. Biol. 11, 236 (2001).
  25. P. Cieplak, F.-Y. Dupradeau, Y. Duan, and J. Wang, Polarization effects in molecular mechanical force fields, J. Phys.: Condens. Matter 21, 333102 (2009).
  26. T. P. Senftle, S. Hong, M. M. Islam, S. B. Kylasa, Y. Zheng, Y. K. Shin, C. Junkermeier, R. Engel-Herbert, M. J. Janik, H. M. Aktulga, T. Verstraelen, A. Grama, and A. C. T. van Duin, The ReaxFF reactive force-field: Development, applications and future directions, npj Comput. Mater. 2, 15011 (2016).
  27. T. W. Ko, J. A. Finkler, S. Goedecker, and J. Behler, General-purpose machine learning potentials capturing nonlocal charge transfer, Acc. Chem. Res. 54, 808 (2021).
  28. J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, and G. E. Dahl, Neural message passing for quantum chemistry, in Proceedings of the 34th International Conference on Machine Learning (ICML 2017) (PMLR, 2017), pp. 1263–1272.
  29. K. T. Schütt, P.-J. Kindermans, H. E. Sauceda, S. Chmiela, A. Tkatchenko, and K.-R. Müller, SchNet: A continuous-filter convolutional neural network for modeling quantum interactions, in Advances in Neural Information Processing Systems 30 (NIPS 2017) (Curran Associates, Red Hook, NY, 2017), pp. 992–1002.
  30. K. T. Schütt, H. E. Sauceda, P.-J. Kindermans, A. Tkatchenko, and K.-R. Müller, SchNet—A deep learning architecture for molecules and materials, J. Chem. Phys. 148, 241722 (2018).
  31. O. T. Unke and M. Meuwly, PhysNet: A neural network for predicting energies, forces, dipole moments, and partial charges, J. Chem. Theory Comput. 15, 3678 (2019).
  32. J. Klicpera, J. Groß, and S. Günnemann, Directional message passing for molecular graphs, in Proceedings of the Eighth International Conference on Learning Representations (ICLR 2020) (OpenReview, 2020).
  33. J. Klicpera, F. Becker, and S. Günnemann, GemNet: Universal directional graph neural networks for molecules, in Advances in Neural Information Processing Systems 34 (NeurIPS 2021) (Curran Associates, Red Hook, NY, 2021), pp. 6790–6802.
  34. O. T. Unke, S. Chmiela, M. Gastegger, K. T. Schütt, H. E. Sauceda, and K.-R. Müller, SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects, Nat. Commun. 12, 7273 (2021).
  35. S. Batzner, A. Musaelian, L. Sun, M. Geiger, J. P. Mailoa, M. Kornbluth, N. Molinari, T. E. Smidt, and B. Kozinsky, E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials, Nat. Commun. 13, 2453 (2022).
  36. I. Batatia, D. P. Kovács, G. N. C. Simm, C. Ortner, and G. Csányi, MACE: Higher order equivariant message passing neural networks for fast and accurate force fields, in Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (Curran Associates, Red Hook, NY, 2022), pp. 11423–11436.
  37. I. Batatia, S. Batzner, D. P. Kovács, A. Musaelian, G. N. C. Simm, R. Drautz, C. Ortner, B. Kozinsky, and G. Csányi, The design space of E(3)-equivariant atom-centered interatomic potentials, arXiv:2205.06643.
  38. A. Bochkarev, Y. Lysogorskiy, C. Ortner, G. Csányi, and R. Drautz, Multilayer atomic cluster expansion for semilocal interactions, Phys. Rev. Res. 4, L042019 (2022).
  39. T. Frank, O. Unke, and K.-R. Müller, So3krates: Equivariant attention for interactions on arbitrary length-scales in molecular systems, in Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (Curran Associates, Red Hook, NY, 2022), pp. 29400–29413.
  40. S. Stocker, J. Gasteiger, F. Becker, S. Günnemann, and J. T. Margraf, How robust are modern graph neural network potentials in long and hot molecular dynamics simulations? Mach. Learn.: Sci. Technol. 3, 045010 (2022).
  41. O. T. Unke, M. Stöhr, S. Ganscha, T. Unterthiner, H. Maennel, S. Kashubin, D. Ahlin, M. Gastegger, L. Medrano Sandonas, A. Tkatchenko, and K.-R. Müller, Accurate machine learned quantum-mechanical force fields for biomolecular simulations, arXiv:2205.08306.
  42. C. Chen and S. P. Ong, A universal graph deep learning interatomic potential for the periodic table, Nat. Comput. Sci. 2, 718 (2022).
  43. G. C. Sosso, D. Donadio, S. Caravati, J. Behler, and M. Bernasconi, Thermal transport in phase-change materials from atomistic simulations, Phys. Rev. B 86, 104301 (2012).
  44. C. Mangold, S. Chen, G. Barbalinardo, J. Behler, P. Pochet, K. Termentzidis, Y. Han, L. Chaput, D. Lacroix, and D. Donadio, Transferability of neural network potentials for varying stoichiometry: Phonons and thermal conductivity of MnxGey compounds, J. Appl. Phys. 127, 244901 (2020).
  45. P. Korotaev, I. Novoselov, A. Yanilkin, and A. Shapeev, Accessing thermal conductivity of complex compounds by machine learning interatomic potentials, Phys. Rev. B 100, 144308 (2019).
  46. H. Liu, X. Qian, H. Bao, C. Zhao, and X. Gu, High-temperature phonon transport properties of SnSe from machine-learning interatomic potential, J. Phys.: Condens. Matter 33, 405401 (2021).
  47. R. Li, E. Lee, and T. Luo, A unified deep neural network potential capable of predicting thermal conductivity of silicon in different phases, Mater. Today Phys. 12, 100181 (2020).
  48. R. Li, Z. Liu, A. Rohskopf, K. Gordiz, A. Henry, E. Lee, and T. Luo, A deep neural network interatomic potential for studying thermal conductivity of β−Ga2O3, Appl. Phys. Lett. 117, 152102 (2020).
  49. X. Qian, S. Peng, X. Li, Y. Wei, and R. Yang, Thermal conductivity modeling using machine learning potentials: Application to crystalline and amorphous silicon, Mater. Today Phys. 10, 100140 (2019).
  50. C. Verdi, F. Karsai, P. Liu, R. Jinnouchi, and G. Kresse, Thermal transport and phase transitions of zirconia by on-the-fly machine-learned interatomic potentials, npj Comput. Mater. 7, 156 (2021).
  51. A. Griewank and A. Walther, Evaluating Derivatives (Society for Industrial and Applied Mathematics, Philadelphia, PA, 2008).
  52. A. G. Baydin, B. A. Pearlmutter, A. A. Radul, and J. M. Siskind, Automatic differentiation in machine learning: A survey, J. Mach. Learn. Res. 18, 1 (2018).
  53. S. Fabris, A. T. Paxton, and M. W. Finnis, Free energy and molecular dynamics calculations for the cubic-tetragonal phase transition in zirconia, Phys. Rev. B 63, 094101 (2001).
  54. C. Carbogno, C. G. Levi, C. G. Van de Walle, and M. Scheffler, Ferroelastic switching of doped zirconia: Modeling and understanding from first principles, Phys. Rev. B 90, 144109 (2014).
  55. R. J. Hardy, Energy-flux operator for a lattice, Phys. Rev. 132, 168 (1963).
  56. See Supplemental Material at http://link.aps.org/supplemental/10.1103/PhysRevB.108.L100302 for a brief introduction to AD and MPNNs; the derivation of (2); discussion of heat flux in solids; additional details on the derivation of Eqs. (5), (7), and (8); details of the implementation of (8); details of hyperparameter selection, training, and testing of the potential, including test set errors, phonon band structures, and the vibrational density of states; convergence of κ with simulation cell size, simulation duration, and noise removal settings; and details of the Green-Kubo workflow. It contains Refs. [79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98].
  57. Y. Chen, Local stress and heat flux in atomistic systems involving three-body forces, J. Chem. Phys. 124, 054113 (2006).
  58. N. C. Admal and E. B. Tadmor, Stress and heat flux for arbitrary multibody potentials: A unified framework, J. Chem. Phys. 134, 184106 (2011).
  59. D. Torii, T. Nakano, and T. Ohara, Contribution of inter- and intramolecular energy transfers to heat conduction in liquids, J. Chem. Phys. 128, 044504 (2008).
  60. Z. Fan, L. F. C. Pereira, H.-Q. Wang, J.-C. Zheng, D. Donadio, and A. Harju, Force and heat current formulas for many-body potentials in molecular dynamics simulations with applications to thermal conductivity calculations, Phys. Rev. B 92, 094301 (2015).
  61. D. Surblys, H. Matsubara, G. Kikugawa, and T. Ohara, Application of atomic stress to compute heat flux via molecular dynamics for systems with many-body interactions, Phys. Rev. E 99, 051301 (2019).
  62. P. Boone, H. Babaei, and C. E. Wilmer, Heat flux for many-body interactions: Corrections to LAMMPS, J. Chem. Theory Comput. 15, 5579 (2019).
  63. M. F. Langer, J. T. Frank, and F. Knoop, Stress and heat flux via automatic differentiation, arXiv:2305.01401.
  64. A. P. Thompson, S. J. Plimpton, and W. Mattson, General formulation of pressure and stress tensor for arbitrary many-body interaction potentials under periodic boundary conditions, J. Chem. Phys. 131, 154107 (2009).
  65. A. Kinaci, J. B. Haskins, and T. Çağın, On calculation of thermal conductivity from Einstein relation in equilibrium molecular dynamics, J. Chem. Phys. 137, 014106 (2012).
  66. S. Raghavan, H. Wang, R. B. Dinwiddie, W. D. Porter, and M. J. Mayo, The effect of grain size, porosity and yttria content on the thermal conductivity of nanocrystalline zirconia, Scr. Mater. 39, 1119 (1998).
  67. J.-F. Bisson, D. Fournier, M. Poulain, O. Lavigne, and R. Mévrel, Thermal conductivity of yttria-zirconia single crystals, determined with spatially resolved infrared thermography, J. Am. Ceram. Soc. 83, 1993 (2000).
  68. R. Mévrel, J.-C. Laizet, A. Azzopardi, B. Leclercq, M. Poulain, O. Lavigne, and D. Demange, Thermal diffusivity and conductivity of Zr1−xYxO2−x/2 (x=0, 0.084 and 0.179) single crystals, J. Eur. Ceram. Soc. 24, 3081 (2004).
  69. V. Blum, R. Gehrke, F. Hanke, P. Havu, V. Havu, X. Ren, K. Reuter, and M. Scheffler, Ab initio molecular simulations with numeric atom-centered orbitals, Comput. Phys. Commun. 180, 2175 (2009).
  70. F. Knoop, T.A.R. Purcell, M. Scheffler, and C. Carbogno, FHI-vibes: Ab initio vibrational simulations, J. Open Source Software 5, 2671 (2020).
  71. J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A. Constantin, X. Zhou, and K. Burke, Restoring the Density-Gradient Expansion for Exchange in Solids and Surfaces, Phys. Rev. Lett. 100, 136406 (2008).
  72. K. T. Schütt, P. Kessel, M. Gastegger, K. A. Nicoli, A. Tkatchenko, and K.-R. Müller, SchNetPack: A deep learning toolbox for atomistic systems, J. Chem. Theory Comput. 15, 448 (2019).
  73. K. Tolborg and A. Walsh, Exploring the high-temperature stabilization of cubic zirconia from anharmonic lattice dynamics, Cryst. Growth Des. 23, 3314 (2023).
  74. R. Patil and E. Subbarao, Axial thermal expansion of ZrO2 and HfO2 in the range room temperature to 1400∘C, J. Appl. Crystallogr. 2, 281 (1969).
  75. E. H. Kisi and C. Howard, Crystal structures of zirconia phases and their inter-relation, Key Eng. Mater. 153–154, 1 (1998).
  76. M. F. Langer, F. Knoop, C. Carbogno, M. Scheffler, and M. Rupp, Data and software: Heat flux for semi-local machine-learning potentials, Zenodo (2023), https://doi.org/10.5281/zenodo.7767432.
  77. https://doi.org/10.17172/NOMAD/2023.03.24-2.
  78. https://marcel.science/gknet.
  79. F. Knuth, C. Carbogno, V. Atalla, V. Blum, and M. Scheffler, All-electron formalism for total energy strain derivatives and stress tensor components for numeric atom-centered orbitals, Comput. Phys. Commun. 190, 33 (2015).
  80. N. Thomas, T. Smidt, S. Kearnes, L. Yang, L. Li, K. Kohlhoff, and P. Riley, Tensor field networks: Rotation- and translation-equivariant neural networks for 3d point clouds, arXiv:1802.08219.
  81. B. Anderson, T.-S. Hy, and R. Kondor, Cormorant: Covariant molecular neural networks, in Advances in Neural Information Processing Systems 32 (NeurIPS 2019) (Curran Associates, Red Hook, NY, 2019), pp. 14537–14546.
  82. K. T. Schütt, O. T. Unke, and M. Gastegger, Equivariant message passing for the prediction of tensorial properties and molecular spectra, in Proceedings of the 38th International Conference on Machine Learning (ICML 2021), pp. 9377–9388.
  83. J. J. Erpenbeck, Einstein-Kubo-Helfand and McQuarrie relations for transport coefficients, Phys. Rev. E 51, 4296 (1995).
  84. R. J. Hardy, Formulas for determining local properties in molecular-dynamics simulations: Shock waves, J. Chem. Phys. 76, 622 (1982).
  85. W. Noll, Die Herleitung der Grundgleichungen der Thermomechanik der Kontinua aus der statistischen Mechanik, J. Ration. Mech. Anal. 4, 627 (1955).
  86. J. H. Irving and J. G. Kirkwood, The statistical mechanical theory of transport processes. IV. The equations of hydrodynamics, J. Chem. Phys. 18, 817 (1950).
  87. A. Ito and H. Nakamura, Energy current on multi-body potential with Dirac delta function, Prog. Theor. Phys. Suppl. 178, 107 (2009).
  88. N. Chetty and R. M. Martin, First-principles energy density and its applications to selected polar surfaces, Phys. Rev. B 45, 6074 (1992).
  89. A. J. Ladd, B. Moran, and W. G. Hoover, Lattice thermal conductivity: A comparison of molecular dynamics and anharmonic lattice dynamics, Phys. Rev. B 34, 5058 (1986).
  90. L. Isaeva, G. Barbalinardo, D. Donadio, and S. Baroni, Modeling heat transport in crystals and glasses from a unified lattice-dynamical approach, Nat. Commun. 10, 3853 (2019).
  91. A. McGaughey and M. Kaviany, Phonon transport in molecular dynamics simulations: Formulation and thermal conductivity prediction, Adv. Heat Transfer 39, 169 (2006).
  92. R. Vogelsang, C. Hoheisel, and G. Ciccotti, Thermal conductivity of the Lennard-Jones liquid by molecular dynamics calculations, J. Chem. Phys. 86, 6371 (1987).
  93. P. C. Howell, Comparison of molecular dynamics methods and interatomic potentials for calculating the thermal conductivity of silicon, J. Chem. Phys. 137, 224111 (2012).
  94. Data set PRL_2017_aiGK in the NOMAD repository, https://doi.org/10.17172/NOMAD/2017.04.13-1.
  95. C. Draxl and M. Scheffler, The NOMAD laboratory: From data sharing to artificial intelligence, J. Phys. Mater. 2, 036001 (2019).
  96. D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, arXiv:1412.6980.
  97. G. J. Martyna, D. J. Tobias, and M. L. Klein, Constant pressure molecular dynamics algorithms, J. Chem. Phys. 101, 4177 (1994).
  98. G. Bussi, D. Donadio, and M. Parrinello, Canonical sampling through velocity rescaling, J. Chem. Phys. 126, 014101 (2007).

Outline

Information

Phys. Rev. B 108, L100302– Published 13 September, 2023
  • Received 30 March 2023
  • Accepted 14 July 2023
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI. Open access publication funded by the Max Planck Society.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation